Data Assimilation of 10 GHz Brightness Temperatures into a Land Surface Model
Jared K. Entin, Paul R. Houser, Jeffrey P. Walker and Eleanor Bourke
In an effort to improve LSM forecasts of soil moisture, data assimilation of various remotely sensed observations has been suggested. The Tropical Rainfall Measuring Mission (TRMM) satellite provides 10 GHz Brightness Temperatures that have a spatial resolution of approximately 45 kilometers. Although not an optimum channel for soil moisture measurement, in part because of possible interference from vegetation, this TRMM signal is the best space borne option currently available. To assimilate the TRMM Brightness Temperature observations using the Kalman filter, a forward radiative transform model has been added to the LSMs run under the Land Data Assimilation Systems (LDAS) project (see http://ldas.gsfc.nasa.gov for more information about LDAS). For this investigation, we use the Common Land Model (CLM) and Mosaic as our LSMs.
The assimilation of TRMM data will be tested for the Oklahoma area, due to the availability of both ground based soil moisture and atmospheric forcing data. However, our initial assimilation studies show the results from synthetic experiments, which use the brightness temperatures generated by our LSMs. In this manner, we assess the potential of using “perfect” 10 GHz TRMM observations to improve our models’ performance when using non-perfect atmospheric forcing and/or initial conditions.